sharing-skills

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Prompt Injection (SAFE): No instructions to bypass safety filters or override system prompts were found.
  • Data Exposure & Exfiltration (SAFE): The skill uses standard Git commands (git push) and the GitHub CLI (gh pr create) to share code. No sensitive file access or hardcoded credentials were detected.
  • Remote Code Execution (SAFE): No remote scripts or unverifiable packages are downloaded or executed.
  • Command Execution (SAFE): The provided bash commands are standard development workflows for branching, committing, and pushing code.
  • Obfuscation (SAFE): No hidden or encoded content was found.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:25 PM